Abstract:
Quantum computing is progressing at a rapid pace, although still constrained by the limitations of noisy intermediate-scale quantum (NISQ) devices, such as restricted qubit counts and high susceptibility to noise. To address these constraints, researchers have begun adapting classical software engineering principles to the quantum realm, giving rise to the field of quantum software engineering. Within this context, techniques, such as circuit-cutting and shotwise distribution, have emerged as promising approaches to divide quantum circuits into smaller fragments and to improve the scalability and reliability of quantum computations. In this article, we present our Cut&Shoot pipeline, which supports the cutting of quantum circuits and the flexible distribution of their shots across multiple quantum processing units (QPUs), in a unified execution strategy. We compare different policies to determine the number of shots for which each fragment should be executed on each QPU. Through 107 000+ experiments on simulated NISQ devices, we assess how circuit size, shot allocation, and QPU availability affect execution times and error rates. Our findings indicate that combining circuit-cutting and shotwise distribution provides a useful tradeoff of execution times and error rates, while featuring high-qubit quantum computations as well as increased resilience and adaptiveness.